In vitro activities do not always translate into in vivo activities, and this is true in Mycobacterium tuberculosis (Mtb) therapeutics. For example, the two first line anti-Mtb drugs isoniazid and rifampicin kill Mtb rapidly in vitro, but their steriliation abilities are slowed and reduced in vivo. Heterogeneity of Mtb populations imparts varied susceptibility to the drugs. In particular, slowly-replicating or non-replicating (collectively NR) Mtb is non-heritably resistant to most first line anti-TB drugs. Our long-term goal is to develop anti-Mtb drugs that kill NR Mtb populations to complement drugs that kill replicating Mtb populations. During the past eight years, components of a prokaryotic ubiquitin-like protein (Pup)-proteasome system have been discovered. Although the Mtb proteasome is dispensable under standard growth conditions, genetic evidence demonstrates its essentiality for Mtb to survive in mice. We also established the concept that despite the essential role of the proteasome in mammals, small chemical molecules can be discovered with extensive (>1000-fold) species selectivity for inhibiting the Mtb proteasome over the human proteasome, and that such inhibition leads to killing of NR Mtb. In other bacteria, disruption of another regulated protein degradation machine, Clp, has led to killing. Thus, either inhibition or forced activation f chambered proteases may represent a new anti-infective strategy. In our recent screening of 1600 capped dipeptides, we identified a lead compound that potently and species-selectively inhibited the Mtb proteasome over the human proteasome. The inhibitors were bactericidal for non-replicating Mtb in synergy with nitric oxide. Using a competition assay, I confirmed that the dipeptides penetrate the mycobacteria and inhibit the proteasome within them. In this application, we will expand our lead compounds into a small focused compound library whose design and synthesis are guided by substrate-profiling, structural analysis, mycobactericidal activity, and metabolic stability.